Table Batched Embedding Operators

std::tuple<at::Tensor, at::Tensor, std::optional<at::Tensor>> get_unique_indices_cuda(const at::Tensor &linear_indices, const int64_t max_indices, const bool compute_count)

Deduplicate indices.

std::tuple<at::Tensor, at::Tensor, std::optional<at::Tensor>, std::optional<at::Tensor>> get_unique_indices_with_inverse_cuda(const at::Tensor &linear_indices, const int64_t max_indices, const bool compute_count, const bool compute_inverse_indices)

Deduplicate indices.

std::tuple<at::Tensor, at::Tensor, std::optional<at::Tensor>> lru_cache_find_uncached_cuda(at::Tensor unique_indices, at::Tensor unique_indices_length, int64_t max_indices, at::Tensor lxu_cache_state, int64_t time_stamp, at::Tensor lru_state, bool gather_cache_stats, at::Tensor uvm_cache_stats, bool lock_cache_line, at::Tensor lxu_cache_locking_counter, const bool compute_inverse_indices)

Lookup LRU cache to find uncached indices, and then sort them based on the set.

int64_t host_lxu_cache_slot(int64_t h_in, int64_t C)

Map index to cache_set. h_in: linear_indices; C: #cache_sets.

at::Tensor linearize_cache_indices_cuda(const at::Tensor &cache_hash_size_cumsum, const at::Tensor &indices, const at::Tensor &offsets, const std::optional<at::Tensor> &B_offsets, const int64_t max_B, const int64_t indices_base_offset)

Linearize the indices of all tables to make it be unique

at::Tensor linearize_cache_indices_from_row_idx_cuda(at::Tensor cache_hash_size_cumsum, at::Tensor update_table_indices, at::Tensor update_row_indices)

Linearize the indices of all tables to make it be unique. Note the update_table_indices and update_row_indices are from the row indices format for inplace update.

at::Tensor direct_mapped_lxu_cache_lookup_cuda(at::Tensor linear_cache_indices, at::Tensor lxu_cache_state, int64_t invalid_index, bool gather_cache_stats, std::optional<at::Tensor> uvm_cache_stats)

LRU cache: fetch the rows corresponding to linear_cache_indices from weights, and insert them into the cache at timestep time_stamp

. void lru_cache_populate_cuda(

at::Tensor weights,

at::Tensor hash_size_cumsum,

int64_t total_cache_hash_size,

at::Tensor cache_index_table_map,

at::Tensor weights_offsets,

at::Tensor D_offsets,

at::Tensor linear_cache_indices,

at::Tensor lxu_cache_state,

at::Tensor lxu_cache_weights,

int64_t time_stamp,

at::Tensor lru_state,

bool stochastic_rounding,

bool gather_cache_stats,

std::optional<at::Tensor> uvm_cache_stats,

bool lock_cache_line,

std::optional<at::Tensor> lxu_cache_locking_counter);

/ / LRU cache: fetch the rows corresponding to linear_cache_indices from /weights, and insert them into the cache at timestep time_stamp

. / weights and lxu_cache_weights have “uint8_t” byte elements void lru_cache_populate_byte_cuda(

at::Tensor weights,

at::Tensor hash_size_cumsum,

int64_t total_cache_hash_size,

at::Tensor cache_index_table_map,

at::Tensor weights_offsets,

at::Tensor weights_tys,

at::Tensor D_offsets,

at::Tensor linear_cache_indices,

at::Tensor lxu_cache_state,

at::Tensor lxu_cache_weights,

int64_t time_stamp,

at::Tensor lru_state,

int64_t row_alignment,

bool gather_cache_stats,

std::optional<at::Tensor> uvm_cache_stats);

/ / Direct-mapped (assoc=1) variant of lru_cache_populate_byte_cuda void direct_mapped_lru_cache_populate_byte_cuda(

at::Tensor weights,

at::Tensor hash_size_cumsum,

int64_t total_cache_hash_size,

at::Tensor cache_index_table_map,

at::Tensor weights_offsets,

at::Tensor weights_tys,

at::Tensor D_offsets,

at::Tensor linear_cache_indices,

at::Tensor lxu_cache_state,

at::Tensor lxu_cache_weights,

int64_t time_stamp,

at::Tensor lru_state,

at::Tensor lxu_cache_miss_timestamp,

int64_t row_alignment,

bool gather_cache_stats,

std::optional<at::Tensor> uvm_cache_stats);

/ / LFU cache: fetch the rows corresponding to linear_cache_indices from /weights

, and insert them into the cache. void lfu_cache_populate_cuda(

at::Tensor weights,

at::Tensor cache_hash_size_cumsum,

int64_t total_cache_hash_size,

at::Tensor cache_index_table_map,

at::Tensor weights_offsets,

at::Tensor D_offsets,

at::Tensor linear_cache_indices,

at::Tensor lxu_cache_state,

at::Tensor lxu_cache_weights,

at::Tensor lfu_state,

bool stochastic_rounding);

/ / LFU cache: fetch the rows corresponding to linear_cache_indices from /weights

, and insert them into the cache. / weights and lxu_cache_weights have “uint8_t” byte elements void lfu_cache_populate_byte_cuda(

at::Tensor weights,

at::Tensor cache_hash_size_cumsum,

int64_t total_cache_hash_size,

at::Tensor cache_index_table_map,

at::Tensor weights_offsets,

at::Tensor weights_tys,

at::Tensor D_offsets,

at::Tensor linear_cache_indices,

at::Tensor lxu_cache_state,

at::Tensor lxu_cache_weights,

at::Tensor lfu_state,

int64_t row_alignment);

/ / Lookup the LRU/LFU cache: find the cache weights location for all indices. / Look up the slots in the cache corresponding to linear_cache_indices

, with / a sentinel value for missing. at::Tensor lxu_cache_lookup_cuda(

at::Tensor linear_cache_indices,

at::Tensor lxu_cache_state,

int64_t invalid_index,

bool gather_cache_stats,

std::optional<at::Tensor> uvm_cache_stats,

std::optional<at::Tensor> num_uniq_cache_indices,

std::optional<at::Tensor> lxu_cache_locations_output);

at::Tensor emulate_cache_miss(

at::Tensor lxu_cache_locations,

const int64_t enforced_misses_per_256,

const bool gather_cache_stats,

at::Tensor uvm_cache_stats);

/ / Lookup the LRU/LFU cache: find the cache weights location for all indices. / Look up the slots in the cache corresponding to linear_cache_indices, with a sentinel value for missing.

void lxu_cache_flush_cuda(at::Tensor uvm_weights, at::Tensor cache_hash_size_cumsum, at::Tensor cache_index_table_map, at::Tensor weights_offsets, at::Tensor D_offsets, int64_t total_D, at::Tensor lxu_cache_state, at::Tensor lxu_cache_weights, bool stochastic_rounding)

Flush the cache: store the weights from the cache to the backing storage.

void reset_weight_momentum_cuda(at::Tensor dev_weights, at::Tensor uvm_weights, at::Tensor lxu_cache_weights, at::Tensor weights_placements, at::Tensor weights_offsets, at::Tensor momentum1_dev, at::Tensor momentum1_uvm, at::Tensor momentum1_placements, at::Tensor momentum1_offsets, at::Tensor D_offsets, at::Tensor pruned_indices, at::Tensor pruned_indices_offsets, at::Tensor logical_table_ids, at::Tensor buffer_ids, at::Tensor cache_hash_size_cumsum, at::Tensor lxu_cache_state, int64_t total_cache_hash_size)

void lxu_cache_locking_counter_decrement_cuda(at::Tensor lxu_cache_locking_counter, at::Tensor lxu_cache_locations)

Decrement the LRU/LFU cache counter based on lxu_cache_locations.

void lxu_cache_locations_update_cuda(at::Tensor lxu_cache_locations, at::Tensor lxu_cache_locations_new, std::optional<at::Tensor> num_uniq_cache_indices)

Inplace update lxu_cache_locations to the new one should only update if lxu_cache_locations[i] == -1 and lxu_cache_locations_new[i] >= 0